The energetically passive behavior of typical prostheses limits the extent to which such prostheses can replicate healthy limb function. Despite this, unilateral transfemoral amputees can largely offset these deficiencies with compensatory actions from their unaffected leg. In the case of bilateral transfemoral amputees, however, the biomechanical deficiencies of passive prostheses cannot be masked by such compensatory actions, and therefore the deficiencies in locomotion are considerably more evident. As a result, although bilateral trans- femoral amputees constitute a relative small proportion of lower extremity amputees, this population is substantially more disabled than their unilateral or transtibial counterparts. A powered prosthesis (i.e., with powered knee and ankle joints) is able to emulate the biomechanical behaviors of the healthy limb to a much greater extent than a passive prosthesis. As such, the underlying hypothesis of this proposal is that a pair of powered prostheses will better restore gait to bilateral transfemoral amputees, relative to passive prostheses. In particular, this hypothesis will be evaluated on six bilateral transfemoral amputee subjects by assessing the extent to which the powered prostheses reduce hip joint torque and power, and the extent to which powered prostheses reduce the metabolic cost of transport, relative to each subject's daily-use passive prostheses.